GitHub-Proxy/eigen
1
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2025-09-12 17:33:15 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix selfadjoint_matrix_vector_product for complex with packet size > 2 (e.g., AVX)

Browse Source
This commit is contained in:
Gael Guennebaud 2014-03-07 23:18:20 +01:00
parent 72461be962
commit ba2f79e680
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
Eigen/src/Core/products/SelfadjointMatrixVector.h
View File

@ -113,9 +113,9 @@ EIGEN_DONT_INLINE void selfadjoint_matrix_vector_product<Scalar,Index,StorageOrd
for (size_t i=starti; i<alignedStart; ++i) for (size_t i=starti; i<alignedStart; ++i)
{ {
res[i] += t0 * A0[i] + t1 * A1[i]; res[i] += cj0.pmul(A0[i], t0) + cj0.pmul(A1[i],t1);
t2 += numext::conj(A0[i]) * rhs[i]; t2 += cj1.pmul(A0[i], rhs[i]);
t3 += numext::conj(A1[i]) * rhs[i]; t3 += cj1.pmul(A1[i], rhs[i]);
} }
// Yes this an optimization for gcc 4.3 and 4.4 (=> huge speed up) // Yes this an optimization for gcc 4.3 and 4.4 (=> huge speed up)
// gcc 4.2 does this optimization automatically. // gcc 4.2 does this optimization automatically.